End of training
Browse files- README.md +4 -3
- all_results.json +12 -0
- eval_results.json +8 -0
- train_results.json +7 -0
- trainer_state.json +624 -0
README.md
CHANGED
|
@@ -2,6 +2,7 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: facebook/wav2vec2-base
|
| 4 |
tags:
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
metrics:
|
| 7 |
- accuracy
|
|
@@ -15,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# wav2vec2-object
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
-
- Accuracy: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: facebook/wav2vec2-base
|
| 4 |
tags:
|
| 5 |
+
- audio-classification
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- accuracy
|
|
|
|
| 16 |
|
| 17 |
# wav2vec2-object
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the MatsRooth/object dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.1285
|
| 22 |
+
- Accuracy: 0.9836
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9836411609498681,
|
| 4 |
+
"eval_loss": 0.12848743796348572,
|
| 5 |
+
"eval_runtime": 10.4681,
|
| 6 |
+
"eval_samples_per_second": 181.026,
|
| 7 |
+
"eval_steps_per_second": 5.732,
|
| 8 |
+
"train_loss": 0.5051058745100385,
|
| 9 |
+
"train_runtime": 1169.0466,
|
| 10 |
+
"train_samples_per_second": 91.81,
|
| 11 |
+
"train_steps_per_second": 0.719
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9836411609498681,
|
| 4 |
+
"eval_loss": 0.12848743796348572,
|
| 5 |
+
"eval_runtime": 10.4681,
|
| 6 |
+
"eval_samples_per_second": 181.026,
|
| 7 |
+
"eval_steps_per_second": 5.732
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"train_loss": 0.5051058745100385,
|
| 4 |
+
"train_runtime": 1169.0466,
|
| 5 |
+
"train_samples_per_second": 91.81,
|
| 6 |
+
"train_steps_per_second": 0.719
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,624 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9836411609498681,
|
| 3 |
+
"best_model_checkpoint": "wav2vec2-object/checkpoint-504",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 840,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.12,
|
| 13 |
+
"learning_rate": 3.5714285714285714e-06,
|
| 14 |
+
"loss": 2.3951,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.24,
|
| 19 |
+
"learning_rate": 7.142857142857143e-06,
|
| 20 |
+
"loss": 2.389,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.36,
|
| 25 |
+
"learning_rate": 1.0714285714285714e-05,
|
| 26 |
+
"loss": 2.3731,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.48,
|
| 31 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 32 |
+
"loss": 2.3439,
|
| 33 |
+
"step": 40
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.6,
|
| 37 |
+
"learning_rate": 1.785714285714286e-05,
|
| 38 |
+
"loss": 2.2776,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.71,
|
| 43 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 44 |
+
"loss": 2.1571,
|
| 45 |
+
"step": 60
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.83,
|
| 49 |
+
"learning_rate": 2.5e-05,
|
| 50 |
+
"loss": 1.92,
|
| 51 |
+
"step": 70
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.95,
|
| 55 |
+
"learning_rate": 2.857142857142857e-05,
|
| 56 |
+
"loss": 1.6367,
|
| 57 |
+
"step": 80
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 1.0,
|
| 61 |
+
"eval_accuracy": 0.933509234828496,
|
| 62 |
+
"eval_loss": 1.284836769104004,
|
| 63 |
+
"eval_runtime": 11.447,
|
| 64 |
+
"eval_samples_per_second": 165.546,
|
| 65 |
+
"eval_steps_per_second": 5.242,
|
| 66 |
+
"step": 84
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.07,
|
| 70 |
+
"learning_rate": 2.9761904761904762e-05,
|
| 71 |
+
"loss": 1.377,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.19,
|
| 76 |
+
"learning_rate": 2.9365079365079366e-05,
|
| 77 |
+
"loss": 1.1976,
|
| 78 |
+
"step": 100
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 1.31,
|
| 82 |
+
"learning_rate": 2.8968253968253967e-05,
|
| 83 |
+
"loss": 1.0806,
|
| 84 |
+
"step": 110
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 1.43,
|
| 88 |
+
"learning_rate": 2.857142857142857e-05,
|
| 89 |
+
"loss": 0.9608,
|
| 90 |
+
"step": 120
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 1.55,
|
| 94 |
+
"learning_rate": 2.8174603174603175e-05,
|
| 95 |
+
"loss": 0.8859,
|
| 96 |
+
"step": 130
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.67,
|
| 100 |
+
"learning_rate": 2.777777777777778e-05,
|
| 101 |
+
"loss": 0.8128,
|
| 102 |
+
"step": 140
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.79,
|
| 106 |
+
"learning_rate": 2.738095238095238e-05,
|
| 107 |
+
"loss": 0.7591,
|
| 108 |
+
"step": 150
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.9,
|
| 112 |
+
"learning_rate": 2.6984126984126984e-05,
|
| 113 |
+
"loss": 0.7071,
|
| 114 |
+
"step": 160
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 2.0,
|
| 118 |
+
"eval_accuracy": 0.9683377308707124,
|
| 119 |
+
"eval_loss": 0.5107668042182922,
|
| 120 |
+
"eval_runtime": 11.2841,
|
| 121 |
+
"eval_samples_per_second": 167.936,
|
| 122 |
+
"eval_steps_per_second": 5.317,
|
| 123 |
+
"step": 168
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 2.02,
|
| 127 |
+
"learning_rate": 2.6587301587301588e-05,
|
| 128 |
+
"loss": 0.6489,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 2.14,
|
| 133 |
+
"learning_rate": 2.6190476190476192e-05,
|
| 134 |
+
"loss": 0.6104,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 2.26,
|
| 139 |
+
"learning_rate": 2.5793650793650793e-05,
|
| 140 |
+
"loss": 0.5735,
|
| 141 |
+
"step": 190
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 2.38,
|
| 145 |
+
"learning_rate": 2.5396825396825397e-05,
|
| 146 |
+
"loss": 0.5251,
|
| 147 |
+
"step": 200
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 2.5,
|
| 151 |
+
"learning_rate": 2.5e-05,
|
| 152 |
+
"loss": 0.4951,
|
| 153 |
+
"step": 210
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.62,
|
| 157 |
+
"learning_rate": 2.4603174603174605e-05,
|
| 158 |
+
"loss": 0.4926,
|
| 159 |
+
"step": 220
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 2.74,
|
| 163 |
+
"learning_rate": 2.4206349206349206e-05,
|
| 164 |
+
"loss": 0.4804,
|
| 165 |
+
"step": 230
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.86,
|
| 169 |
+
"learning_rate": 2.380952380952381e-05,
|
| 170 |
+
"loss": 0.4279,
|
| 171 |
+
"step": 240
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 2.98,
|
| 175 |
+
"learning_rate": 2.3412698412698414e-05,
|
| 176 |
+
"loss": 0.4246,
|
| 177 |
+
"step": 250
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 3.0,
|
| 181 |
+
"eval_accuracy": 0.97467018469657,
|
| 182 |
+
"eval_loss": 0.30182337760925293,
|
| 183 |
+
"eval_runtime": 11.1208,
|
| 184 |
+
"eval_samples_per_second": 170.401,
|
| 185 |
+
"eval_steps_per_second": 5.395,
|
| 186 |
+
"step": 252
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 3.1,
|
| 190 |
+
"learning_rate": 2.301587301587302e-05,
|
| 191 |
+
"loss": 0.3991,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 3.21,
|
| 196 |
+
"learning_rate": 2.261904761904762e-05,
|
| 197 |
+
"loss": 0.381,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 3.33,
|
| 202 |
+
"learning_rate": 2.222222222222222e-05,
|
| 203 |
+
"loss": 0.3748,
|
| 204 |
+
"step": 280
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 3.45,
|
| 208 |
+
"learning_rate": 2.1825396825396827e-05,
|
| 209 |
+
"loss": 0.3619,
|
| 210 |
+
"step": 290
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 3.57,
|
| 214 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 215 |
+
"loss": 0.3416,
|
| 216 |
+
"step": 300
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 3.69,
|
| 220 |
+
"learning_rate": 2.1031746031746032e-05,
|
| 221 |
+
"loss": 0.3261,
|
| 222 |
+
"step": 310
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 3.81,
|
| 226 |
+
"learning_rate": 2.0634920634920633e-05,
|
| 227 |
+
"loss": 0.2987,
|
| 228 |
+
"step": 320
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 3.93,
|
| 232 |
+
"learning_rate": 2.023809523809524e-05,
|
| 233 |
+
"loss": 0.3173,
|
| 234 |
+
"step": 330
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 4.0,
|
| 238 |
+
"eval_accuracy": 0.9767810026385224,
|
| 239 |
+
"eval_loss": 0.2097773551940918,
|
| 240 |
+
"eval_runtime": 10.8229,
|
| 241 |
+
"eval_samples_per_second": 175.091,
|
| 242 |
+
"eval_steps_per_second": 5.544,
|
| 243 |
+
"step": 336
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 4.05,
|
| 247 |
+
"learning_rate": 1.984126984126984e-05,
|
| 248 |
+
"loss": 0.2915,
|
| 249 |
+
"step": 340
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 4.17,
|
| 253 |
+
"learning_rate": 1.9444444444444445e-05,
|
| 254 |
+
"loss": 0.2797,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 4.29,
|
| 259 |
+
"learning_rate": 1.9047619047619046e-05,
|
| 260 |
+
"loss": 0.2736,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 4.4,
|
| 265 |
+
"learning_rate": 1.8650793650793654e-05,
|
| 266 |
+
"loss": 0.285,
|
| 267 |
+
"step": 370
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 4.52,
|
| 271 |
+
"learning_rate": 1.8253968253968254e-05,
|
| 272 |
+
"loss": 0.2456,
|
| 273 |
+
"step": 380
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 4.64,
|
| 277 |
+
"learning_rate": 1.7896825396825394e-05,
|
| 278 |
+
"loss": 0.2508,
|
| 279 |
+
"step": 390
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 4.76,
|
| 283 |
+
"learning_rate": 1.7500000000000002e-05,
|
| 284 |
+
"loss": 0.2649,
|
| 285 |
+
"step": 400
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 4.88,
|
| 289 |
+
"learning_rate": 1.7103174603174603e-05,
|
| 290 |
+
"loss": 0.2643,
|
| 291 |
+
"step": 410
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 5.0,
|
| 295 |
+
"learning_rate": 1.6706349206349207e-05,
|
| 296 |
+
"loss": 0.2585,
|
| 297 |
+
"step": 420
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 5.0,
|
| 301 |
+
"eval_accuracy": 0.9783641160949869,
|
| 302 |
+
"eval_loss": 0.1684747189283371,
|
| 303 |
+
"eval_runtime": 11.0873,
|
| 304 |
+
"eval_samples_per_second": 170.917,
|
| 305 |
+
"eval_steps_per_second": 5.412,
|
| 306 |
+
"step": 420
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 5.12,
|
| 310 |
+
"learning_rate": 1.6309523809523807e-05,
|
| 311 |
+
"loss": 0.2413,
|
| 312 |
+
"step": 430
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 5.24,
|
| 316 |
+
"learning_rate": 1.5912698412698415e-05,
|
| 317 |
+
"loss": 0.2137,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 5.36,
|
| 322 |
+
"learning_rate": 1.5515873015873016e-05,
|
| 323 |
+
"loss": 0.2007,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 5.48,
|
| 328 |
+
"learning_rate": 1.511904761904762e-05,
|
| 329 |
+
"loss": 0.2259,
|
| 330 |
+
"step": 460
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 5.6,
|
| 334 |
+
"learning_rate": 1.4722222222222222e-05,
|
| 335 |
+
"loss": 0.1921,
|
| 336 |
+
"step": 470
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 5.71,
|
| 340 |
+
"learning_rate": 1.4325396825396825e-05,
|
| 341 |
+
"loss": 0.1951,
|
| 342 |
+
"step": 480
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 5.83,
|
| 346 |
+
"learning_rate": 1.3928571428571429e-05,
|
| 347 |
+
"loss": 0.2063,
|
| 348 |
+
"step": 490
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 5.95,
|
| 352 |
+
"learning_rate": 1.3531746031746031e-05,
|
| 353 |
+
"loss": 0.192,
|
| 354 |
+
"step": 500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 6.0,
|
| 358 |
+
"eval_accuracy": 0.9836411609498681,
|
| 359 |
+
"eval_loss": 0.12848743796348572,
|
| 360 |
+
"eval_runtime": 11.2118,
|
| 361 |
+
"eval_samples_per_second": 169.018,
|
| 362 |
+
"eval_steps_per_second": 5.351,
|
| 363 |
+
"step": 504
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 6.07,
|
| 367 |
+
"learning_rate": 1.3134920634920635e-05,
|
| 368 |
+
"loss": 0.2192,
|
| 369 |
+
"step": 510
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 6.19,
|
| 373 |
+
"learning_rate": 1.2738095238095238e-05,
|
| 374 |
+
"loss": 0.2063,
|
| 375 |
+
"step": 520
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 6.31,
|
| 379 |
+
"learning_rate": 1.2341269841269842e-05,
|
| 380 |
+
"loss": 0.2053,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 6.43,
|
| 385 |
+
"learning_rate": 1.1944444444444444e-05,
|
| 386 |
+
"loss": 0.2138,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 6.55,
|
| 391 |
+
"learning_rate": 1.1547619047619048e-05,
|
| 392 |
+
"loss": 0.1777,
|
| 393 |
+
"step": 550
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 6.67,
|
| 397 |
+
"learning_rate": 1.1150793650793651e-05,
|
| 398 |
+
"loss": 0.1833,
|
| 399 |
+
"step": 560
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 6.79,
|
| 403 |
+
"learning_rate": 1.0753968253968255e-05,
|
| 404 |
+
"loss": 0.1803,
|
| 405 |
+
"step": 570
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 6.9,
|
| 409 |
+
"learning_rate": 1.0357142857142857e-05,
|
| 410 |
+
"loss": 0.1772,
|
| 411 |
+
"step": 580
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 7.0,
|
| 415 |
+
"eval_accuracy": 0.9810026385224274,
|
| 416 |
+
"eval_loss": 0.12096092104911804,
|
| 417 |
+
"eval_runtime": 10.7065,
|
| 418 |
+
"eval_samples_per_second": 176.995,
|
| 419 |
+
"eval_steps_per_second": 5.604,
|
| 420 |
+
"step": 588
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 7.02,
|
| 424 |
+
"learning_rate": 9.960317460317462e-06,
|
| 425 |
+
"loss": 0.1887,
|
| 426 |
+
"step": 590
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 7.14,
|
| 430 |
+
"learning_rate": 9.563492063492064e-06,
|
| 431 |
+
"loss": 0.1695,
|
| 432 |
+
"step": 600
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 7.26,
|
| 436 |
+
"learning_rate": 9.166666666666668e-06,
|
| 437 |
+
"loss": 0.1767,
|
| 438 |
+
"step": 610
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 7.38,
|
| 442 |
+
"learning_rate": 8.76984126984127e-06,
|
| 443 |
+
"loss": 0.1694,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 7.5,
|
| 448 |
+
"learning_rate": 8.373015873015875e-06,
|
| 449 |
+
"loss": 0.1722,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 7.62,
|
| 454 |
+
"learning_rate": 7.976190476190477e-06,
|
| 455 |
+
"loss": 0.1525,
|
| 456 |
+
"step": 640
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 7.74,
|
| 460 |
+
"learning_rate": 7.57936507936508e-06,
|
| 461 |
+
"loss": 0.1517,
|
| 462 |
+
"step": 650
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 7.86,
|
| 466 |
+
"learning_rate": 7.182539682539683e-06,
|
| 467 |
+
"loss": 0.1634,
|
| 468 |
+
"step": 660
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 7.98,
|
| 472 |
+
"learning_rate": 6.785714285714286e-06,
|
| 473 |
+
"loss": 0.1435,
|
| 474 |
+
"step": 670
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 8.0,
|
| 478 |
+
"eval_accuracy": 0.9804749340369393,
|
| 479 |
+
"eval_loss": 0.1141170933842659,
|
| 480 |
+
"eval_runtime": 10.8149,
|
| 481 |
+
"eval_samples_per_second": 175.222,
|
| 482 |
+
"eval_steps_per_second": 5.548,
|
| 483 |
+
"step": 672
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 8.1,
|
| 487 |
+
"learning_rate": 6.388888888888889e-06,
|
| 488 |
+
"loss": 0.1874,
|
| 489 |
+
"step": 680
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 8.21,
|
| 493 |
+
"learning_rate": 5.992063492063493e-06,
|
| 494 |
+
"loss": 0.1268,
|
| 495 |
+
"step": 690
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 8.33,
|
| 499 |
+
"learning_rate": 5.595238095238095e-06,
|
| 500 |
+
"loss": 0.1529,
|
| 501 |
+
"step": 700
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 8.45,
|
| 505 |
+
"learning_rate": 5.198412698412698e-06,
|
| 506 |
+
"loss": 0.1356,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 8.57,
|
| 511 |
+
"learning_rate": 4.801587301587302e-06,
|
| 512 |
+
"loss": 0.1549,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 8.69,
|
| 517 |
+
"learning_rate": 4.404761904761905e-06,
|
| 518 |
+
"loss": 0.1775,
|
| 519 |
+
"step": 730
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 8.81,
|
| 523 |
+
"learning_rate": 4.007936507936508e-06,
|
| 524 |
+
"loss": 0.1448,
|
| 525 |
+
"step": 740
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 8.93,
|
| 529 |
+
"learning_rate": 3.611111111111111e-06,
|
| 530 |
+
"loss": 0.1425,
|
| 531 |
+
"step": 750
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 9.0,
|
| 535 |
+
"eval_accuracy": 0.9836411609498681,
|
| 536 |
+
"eval_loss": 0.10937827080488205,
|
| 537 |
+
"eval_runtime": 11.2971,
|
| 538 |
+
"eval_samples_per_second": 167.742,
|
| 539 |
+
"eval_steps_per_second": 5.311,
|
| 540 |
+
"step": 756
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 9.05,
|
| 544 |
+
"learning_rate": 3.2142857142857143e-06,
|
| 545 |
+
"loss": 0.1726,
|
| 546 |
+
"step": 760
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 9.17,
|
| 550 |
+
"learning_rate": 2.8174603174603176e-06,
|
| 551 |
+
"loss": 0.1471,
|
| 552 |
+
"step": 770
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 9.29,
|
| 556 |
+
"learning_rate": 2.4206349206349204e-06,
|
| 557 |
+
"loss": 0.151,
|
| 558 |
+
"step": 780
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 9.4,
|
| 562 |
+
"learning_rate": 2.0238095238095237e-06,
|
| 563 |
+
"loss": 0.1586,
|
| 564 |
+
"step": 790
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 9.52,
|
| 568 |
+
"learning_rate": 1.626984126984127e-06,
|
| 569 |
+
"loss": 0.1112,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 9.64,
|
| 574 |
+
"learning_rate": 1.23015873015873e-06,
|
| 575 |
+
"loss": 0.1315,
|
| 576 |
+
"step": 810
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 9.76,
|
| 580 |
+
"learning_rate": 8.333333333333333e-07,
|
| 581 |
+
"loss": 0.1649,
|
| 582 |
+
"step": 820
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 9.88,
|
| 586 |
+
"learning_rate": 4.365079365079365e-07,
|
| 587 |
+
"loss": 0.1458,
|
| 588 |
+
"step": 830
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 10.0,
|
| 592 |
+
"learning_rate": 3.968253968253968e-08,
|
| 593 |
+
"loss": 0.1398,
|
| 594 |
+
"step": 840
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 10.0,
|
| 598 |
+
"eval_accuracy": 0.9825857519788919,
|
| 599 |
+
"eval_loss": 0.1077321395277977,
|
| 600 |
+
"eval_runtime": 11.3409,
|
| 601 |
+
"eval_samples_per_second": 167.094,
|
| 602 |
+
"eval_steps_per_second": 5.291,
|
| 603 |
+
"step": 840
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 10.0,
|
| 607 |
+
"step": 840,
|
| 608 |
+
"total_flos": 1.170126007537755e+18,
|
| 609 |
+
"train_loss": 0.5051058745100385,
|
| 610 |
+
"train_runtime": 1169.0466,
|
| 611 |
+
"train_samples_per_second": 91.81,
|
| 612 |
+
"train_steps_per_second": 0.719
|
| 613 |
+
}
|
| 614 |
+
],
|
| 615 |
+
"logging_steps": 10,
|
| 616 |
+
"max_steps": 840,
|
| 617 |
+
"num_input_tokens_seen": 0,
|
| 618 |
+
"num_train_epochs": 10,
|
| 619 |
+
"save_steps": 500,
|
| 620 |
+
"total_flos": 1.170126007537755e+18,
|
| 621 |
+
"train_batch_size": 32,
|
| 622 |
+
"trial_name": null,
|
| 623 |
+
"trial_params": null
|
| 624 |
+
}
|